The majority of weather radar systems in operation today utilize a single polarization strategy to enhance precipitation reflectivity. Liquid hydrometeors (e.g. raindrops) deviate from a sphere when their radius is greater than about 1 mm and have a shape more like that of an oblate spheroid with a flattened base (similar to a hamburger bun) that gives a slightly stronger horizontal return. Hence, current radar systems are typically horizontally polarized to enhance precipitation returns.
However, singly polarized radar systems have severe limitations in regions with partial beam blockage and such systems do not facilitate hydrometeor classification. To overcome these shortcomings of singly polarized weather radar systems, systems with alternating pulses of horizontally and vertically polarized signals have been developed. These dual polarized radar system, sometimes referred to as “polarimetric weather radars,” offer several advantages over conventional radars in estimating precipitation types and amounts. Foremost among these advantages are the capability to discriminate between hail and rain, detect mixed phase precipitation, and estimate rainfall volume.
Current dual polarized radar systems utilize polarization that is altered sequentially between linear vertical and linear horizontal to capture data enhancing values, such as, for example: (1) reflectivity factors at both horizontal and vertical polarization; (2) differential reflectivity for two reflectivity factors; (3) cumulative differential phasing between the horizontally and vertically polarized echoes; (4) correlation coefficients between vertically and horizontally polarized echoes; and (5) linear depolarization ratios. In addition, Doppler velocity and spectrum width can be obtained by suitably processing the horizontally and vertically polarized return signals.
Dual polarized radar systems also allow for the implementation of precipitation classification schemes from inference radar processing of hydrometeor shapes as discussed in various papers authored by practitioners who work in these areas, such as, Ryzhkov, Liu, Vivekanandan, and Zrnic. In addition, by looking at phase differences between the horizontal and vertical components, the effects of partial beam blockage can be mitigated and greater clutter rejection can be obtained. However, the underlying assumption is that subsequent pulses (those of each polarization) are highly correlated and provide an effective velocity range reduced by a factor of two.
While dual polarized radar systems provide enhanced resolving of hydrometeor data, current systems typically utilize high power, high speed waveguide switches to transfer the transmitted energy between dual waveguide conduits. The use of waveguide switches to separate orthogonally oriented energy waves has practical limitations that inhibit the widespread implementation and acceptance of dual polarized radar systems for a number of reasons, as will be discussed.
Current dual polarization weather radar systems switch between polarization modes on a pulse by pulse basis as shown in representative system 10 at FIG. 1. To switch between polarization modes, a high power, high speed waveguide switch 17 (referred to in the industry by the generic term “waveguide orthomode switch”) to transfer the transmitted energy between a horizontally oriented waveguide 18 and a vertically oriented waveguide 19. The waveguide phase shift is induced by Faraday rotation, which is a well known technique in the industry. This representative system includes elements known in the art, such as a Klystron based transmitter 11, a bi-directional coupler 12, a channel circulator 13, a digital receiver 14, rotational couplings 15 such as elevational and azimuth joints, and the requisite pedestal mounting 16 with feed horns and antenna.
Such a representative system is similar to the one disclosed by Zrnic in U.S. Pat. No. 5,500,646, except that Zrnic discloses a theoretical simultaneous dual polarized radar system by replacing the orthomode switch with a power splitter and an orthomode coupling at the antenna feed horn. Zrnic also worked out the various calculations pertaining to simultaneous dual polarization radar systems as recorded in the '646 patent, not already known in the industry, and such calculations are hereby incorporated by reference for background information into this disclosure and are applicable for the presently disclosed systems. While Zrnic displays a theoretic simultaneous dual radar system, the presently disclosed systems offer improvements to the Zrnic model to allow for real-world implementation and practical application of the Zrnic model to modern weather radar systems, as will be described. A number of systems exist today for the actual capturing of radar reflectivity data and incorporation into a local workstations or a nation-wide network of radar installations. One such system is disclosed in a white paper authored by A. Zahrai and D. Zrnic entitled Implementation of Polarimetric Capability for the WSR-88D (NEXRAD) Radar, published in American Meteorological Society in 1997 in section 9.5, which is hereby incorporated by reference. Additional comments pertaining to the capturing of reflectivity data and the processing of such data will not be made as these incorporated references describe the basic theory and operation of such systems and such information is already understood in the industry and not necessary for a complete understanding of the herein described invention.
For dual alternating dual polarization systems, a primary practical problem is the limitations encountered in the use of the high power switch 12. These switches are specialized pieces of equipment and tend to be very expensive and difficult to maintain. In addition they tend to exhibit relatively low isolation capability between the two modes. Manufactures familiar with the installation and maintenance of these systems in the field have found the reliability of the high power, dual polarization switches to decrease as the transmission frequency decreases, thereby limiting the practical implementation of dual polarization weather radar systems.
The dual polarization switch 12 is a ferrite based phase shifter. The switch operates by establishing a magnetic field in the ferrite core before the transmission of the pulse. The coupling of the magnetic field and the electromagnetic pulse causes Faraday rotation, i.e. rotation of the plane of polarization as the pulse passes through the ferrite medium. Through this process, the energy from the pulse is directed to one of two output ports, a horizontally oriented port and a vertically oriented port.
The size of the ferrite core is highly dependent upon the wavelength and hence frequency of the radar system. The longer the wavelength, the larger the ferrite core and the greater the doped surface area which will absorb microwave energy. Those skilled in the art currently believe that switch failures in the lower relative frequency rate of S-band (e.g. 10 cm wave lengths) are due to the changes in crystalline structure of the ferrite material resulting from the increased energy absorption at this frequency. At higher frequencies, e.g. C and X bands, manufactures have experienced less failure issues, but still encounter some problems. Other organizations have implemented mechanical rather than electromagnetic switches in S-band systems. However, mechanical switches have other limitations, known in the art, such as a fixed operating frequency for a selected switch assembly, which limits the operational radar system parameters to a fixed pulse interval period.
Another limitation of current alternating dual polarization radar systems is long dwell times and velocity range reductions. Any received reflection signal resulting from either polarization modes is assumed to come from the same scatterers (e.g. hydrometeors). In order to correlate the data from both the horizontally polarized and vertically polarized channels in current systems utilizing a waveguide switch, a single polarization pulse is transmitted followed by a period of delay (the dwell time) while reflections signals are being received. The opposing polarity pulse is subsequently sent and additional data is received by the same (single) receiver chain during a second dwell time. Reception of reflection signals, therefore, occurs during these two dwell periods during antenna rotation within a single beamwidth, resulting in a longer total dwell time for each beamwidth interrogation. Similarly, since the dwell time for each beamwidth interrogation (vertical+horizontal) is doubled, computational velocity perception is halved, thereby limiting the ability of current systems to resolve relatively high wind velocities in radar returns.
Additionally, practical problems exist in the above referenced Zrnic models. Transmission of wave pulse trains over long waveguides introduces phase and amplitude errors in the waveform which can interfere with the processing of radar reflectivity signals. Also, transmitting radar pulse trains through the azimuth and elevational joints introduces additional errors and causes signal attenuation. Such errors require complex compensation processing in the received radar returns limiting the reliability of reflectivity data in simultaneous dual polarization weather radar systems.
Therefore, what is needed is an improvement in dual polarization weather radar systems from the current methods of alternating polarization modes to simultaneous dual polarization modes to solve issues such as long dwell times and velocity range reductions, and to eliminate expensive and hard to maintain dual polarization switches currently employed in dual polarization weather radar systems.